A numerical algorithm for finding solution of sign-indefinite algebraic Riccati equations for general multiparameter singularly perturbed systems (Q2381315)

A linear time-invariant general multiparameter singularly perturbed system is considered. An \(H_\infty\) filter to estimate system states is designed. Like \textit{M.~T.~Lim} and \textit{Z.~Gajić} [``Reduced-order \(H_\infty\) optimal filtering for systems with slow and fast modes'', IEEE Trans. Circ. Syst. I, 47, No. 2, 250--254 (2000)], it yields a sign-indefinite general multiparameter algebraic Riccati equation; in a matrix form: \(AX + XA^T - XSX + U = 0\). In this work such an equation is investigated. The main result is a new algorithm that combines the Newton's method with two fixed point algorithms; the quadratic convergence is shown. The author: ``\dots it is shown that the uniqueness and positive semidefiniteness of the convergence solutions are guaranteed in the neighborhood of the initial conditions.'' Two computational examples are present.